Authors: Shaudys, Fred E.; Leen, Todd K.
Source: Proceedings of the International Joint Conference on Neural Networks, IEEE, Baltimore, 1992
We apply the feature-selection technique of Fukunaga and Koontz, an extension of the Karhunen-Loeve transformation, to spoken letter recognition. Feedforward networks trained for letter-pair discrimination with the new features show up to 37% reduction in classifier error rate relative to networks trained with spectral coefficients. This performance increase is accompanied by a 91% reduction in feature dimension. For three-letter discrimination, the new features perform comparably to spectral coefficients, with a 9\% reduction in feature dimension.
NeuroSolutions Home Page